A Human-LLM Teaming Framework for Privacy Risk Analysis: An Illustration with CBDC-Based Welfare Schemes

2026-08-17

AI Summary

This study addresses the complexity of privacy risk assessment in Central Bank Digital Currency (CBDC) welfare programs by proposing Priam, a human-AI collaborative framework. As the first synergistic analytical framework tailored to this domain, Priam integrates Large Language Models with expert feedback through iterative collaboration to decouple evidence reasoning from information gap identification. By effectively combining AI-driven document processing capabilities with human professional judgment, the framework establishes a novel paradigm for complex privacy compliance analysis. Empirical evaluations demonstrate that Priam significantly outperforms single-agent approaches in data classification and attribute assignment tasks. Consequently, this method substantially enhances both the accuracy and trustworthiness of risk assessments, offering a robust solution for navigating the intricate privacy challenges inherent in CBDC welfare systems.

Abstract

Central Bank Digital Currency (CBDC)-based welfare schemes may be potentially privacy invasive as they process significant volumes of beneficiary personal data and lead to privacy harms such as surveillance, discrimination and stigmatization. Such welfare delivery schemes involve complex digital ecosystems and large number of stakeholders. Consequently, to examine their privacy risks, privacy risk assessments require extensive information gathering and synthesis, complex reasoning, scenario explorations, contextual evaluation and human judgement. Thus, they present ideal scenarios for human-LLM teaming, where effective integration of complementary human and LLM capabilities can yield an outcome far superior to either human-only or LLM-only assessments. In this paper, we propose a first human-LLM teaming framework for the systematic privacy risk analysis methodology called PRIAM. The framework specifies an iterative collaborative process in which the LLM processes large-scale documentary evidence to produce initial outputs, which are then interpreted and evaluated by human experts who direct their further refinement by the LLM and exercise their judgement to finalize the output. We illustrate the framework on the data characterization activity of PRIAM using a CBDC-based welfare scheme use case. The illustration demonstrates that while LLMs generate the initial data categories and assign initial values to data attributes, human experts evaluate and provide feedback to refine them, distinguishing documented evidence from inferences, identifying information gaps, and flagging unsupported or ambiguous outputs. This framework serves as a foundational contribution towards human-AI teaming for privacy risk assessments.

Problem

• Privacy Risk Analysis
• Human-LLM Teaming
• CBDC Welfare Schemes
• Privacy Assessment

Innovation

• Human-LLM Teaming
• Privacy Risk Analysis
• PRIAM Framework
• CBDC Welfare Schemes
• Iterative Collaboration